• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

*Heaven Extreme*

[email protected]

670 GTX SLI @ 1200/1800

00004.jpg


http://i.imgur.com/ZF87y.jpg

6d5.png


http://www.techpowerup.com/gpuz/emmfm/
 
majnu said:
How come these 7950's are performing just as well as GTX680's? :eek:
Click to expand...

680s are a bit gimped on the compute side relative to what should be a high end card and nVidia's tessellation processing depends heavily on that... I do wonder if AMD's optimised tessellation settings have anything to do with it tho - last time I checked I don't think it was working for heaven but its possible it now does and/or forcing it through radeonpro works unfortunatly don't have a card here to test atm. If those settings are working with heaven tho then the AMD results would be invalid. There is another way to cheat the tessellation and dynamically clip it to a level as well (while the extreme setting is set in heaven) but I doubt people are going as far as using a 3rd party application to cheat a heaven benchmark - never know tho someone on another forum was caught out doing it (forgot to revert to normal level when taking the final screenshot).
 
Rroff said:
680s are a bit gimped on the compute side relative to what should be a high end card and nVidia's tessellation processing depends heavily on that... I do wonder if AMD's optimised tessellation settings have anything to do with it tho - last time I checked I don't think it was working for heaven but its possible it now does and/or forcing it through radeonpro works unfortunatly don't have a card here to test atm. If those settings are working with heaven tho then the AMD results would be invalid. There is another way to cheat the tessellation and dynamically clip it to a level as well (while the extreme setting is set in heaven) but I doubt people are going as far as using a 3rd party application to cheat a heaven benchmark - never know tho someone on another forum was caught out doing it (forgot to revert to normal level when taking the final screenshot).
Click to expand...

It is like any kind of hacking, someone will because they can. For me it comes down to personal pride and you get none of that with cheating.
 
Yeah I don't really do heaven benchmarks now because of it (to post online) even on these forums a few people were posting results where they changed the numbers in the HTML report... til someone let on there was a way to validate the numbers (funny how a few results mysteriously dissapeared after that came out).
 
Rroff said:
Yeah I don't really do heaven benchmarks now because of it (to post online) even on these forums a few people were posting results where they changed the numbers in the HTML report... til someone let on there was a way to validate the numbers (funny how a few results mysteriously dissapeared after that came out).
Click to expand...

Yep, I remember that and was very easy to alter.
 
so i'm presuming shaders mean nothing in this test, and its all about out and out core/mem speed, was just comparing Kamz to mine, thats one cracking 7950, for this test anyway
 
Having moved over from Nvidia to ATI i must say there are some different options available to ATI, regarding the AMD's optimised tessellation (not sure if its allowed) but mine seems to be on, furthermore I have 12.11 caps installed and i believe these control settings also? As I tired to make a profile for borderlands 2/3d mark11 etc and it stated another profile already in place, however i have no idea what settings it has as the profiles are not viewable.

3Dmark 2011 picks up Virtua mvp what ever is called (1340 core link below):

http://www.3dmark.com/3dm11/4999631

Kamz
 
^^ I have no idea if AMD's optimised tessellation is doing anything with heaven or not (or even if they have a forced profile at driver level). It didn't used to but I don't think anyones checked any recent drivers. Only way to tell really is either high res wireframe comparisions or low level debugging with something equivalent to perfkit.
 
I also suspect jiggery pokerey with these 12.11 drivers.

It seems very strange that a company in financial trouble suddenly pull this trump card out of the hat.

Time will tell.
 
SkodaMart said:
I also suspect jiggery pokerey with these 12.11 drivers.

It seems very strange that a company in financial trouble suddenly pull this trump card out of the hat.

Time will tell.
Click to expand...

Thing is the performance increases are not just based off the benchmarks but evident in games also, for instance BF3 and others have gained some major long awaited increases.

I believe Rusty went from Nvidia dual 680 to dual 7950 and claimed to have just as good if not better results which im sure it was also down to the 12.11 beta's

Kamz
 
There are a few reports of ATi cards running hotter with 12.11

More fps more heat that sort of makes sense.
 
Well just installed my Gigabyte Windforce 7950 @ stock 900/1250 and ran a few loops of Heaven. Idles at 26-28c @30% fan, hottest it got in Heaven running the extreme settings was 51c @ 0.967v with 12.11 beta 8. So no heat problems there.:p
 
SkodaMart said:
More fps more heat that sort of makes sense.
Click to expand...

Doesn't. Heat is related to the voltage you're putting through it.

kamz said:
I believe Rusty went from Nvidia dual 680 to dual 7950 and claimed to have just as good if not better results which im sure it was also down to the 12.11 beta's

Kamz
Click to expand...

Bit slower overall at normal 1080 (faster depending on the game). Faster if any kind of compute is required. Way faster at triple screen res.
 
You must log in or register to reply here.
Back
Top Bottom